 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I updated my spaceship scene and rendered new frames to be used with
Pannellum panorama viewer.
http://isometricland.net/panorama/pannellum.htm?config=pano-gh-scene-spinner.json
I still was not able to turn the interior atmosphere due to render
times. Even on the 16-core MS Azure server I am using for my renders
this month, things were taking too long.
Enjoy.
Mike
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 4/2/2017 8:52 PM, Mike Horvath wrote:
> I updated my spaceship scene and rendered new frames to be used with
> Pannellum panorama viewer.
>
> http://isometricland.net/panorama/pannellum.htm?config=pano-gh-scene-spinner.json
>
>
> I still was not able to turn the interior atmosphere due to render
> times. Even on the 16-core MS Azure server I am using for my renders
> this month, things were taking too long.
>
> Enjoy.
>
>
> Mike
Forgot to mention there are strange square shaped artifacts in the blue
end caps of the scene. They exist in the source images as well. I don't
know what causes them.
Mike
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Mike Horvath <mik### [at] gmail com> wrote:
> On 4/2/2017 8:52 PM, Mike Horvath wrote:
> > I updated my spaceship scene and rendered new frames to be used with
> > Pannellum panorama viewer.
> >
> > http://isometricland.net/panorama/pannellum.htm?config=pano-gh-scene-spinner.json
> >
> >
> > I still was not able to turn the interior atmosphere due to render
> > times. Even on the 16-core MS Azure server I am using for my renders
> > this month, things were taking too long.
>
> Forgot to mention there are strange square shaped artifacts in the blue
> end caps of the scene. They exist in the source images as well. I don't
> know what causes them.
Impressive scene. And maybe disorienting to some earthbound people. :)
If able to see it, is the atmosphere both ground-hugging and cloudlets floating
around?
Wild guess about the squares. Could it be something about the ends being
directly opposite and slightly reflective, max_trace_level?
Although I don't recall seeing unusual squared patterns from anything like that
before.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 4/3/2017 9:50 AM, omniverse wrote:
> Mike Horvath <mik### [at] gmailcom> wrote:
>> On 4/2/2017 8:52 PM, Mike Horvath wrote:
>>> I updated my spaceship scene and rendered new frames to be used with
>>> Pannellum panorama viewer.
>>>
>>> http://isometricland.net/panorama/pannellum.htm?config=pano-gh-scene-spinner.json
>>>
>>>
>>> I still was not able to turn the interior atmosphere due to render
>>> times. Even on the 16-core MS Azure server I am using for my renders
>>> this month, things were taking too long.
>>
>> Forgot to mention there are strange square shaped artifacts in the blue
>> end caps of the scene. They exist in the source images as well. I don't
>> know what causes them.
>
> Impressive scene. And maybe disorienting to some earthbound people. :)
>
Yes it is impressive. :)
> If able to see it, is the atmosphere both ground-hugging and cloudlets floating
> around?
>
> Wild guess about the squares. Could it be something about the ends being
> directly opposite and slightly reflective, max_trace_level?
> Although I don't recall seeing unusual squared patterns from anything like that
> before.
>
I would definitely check the max level first.
When you were making the trefoil, is there anything with an unusual
normal involved?
--
Regards
Stephen
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 03.04.2017 um 03:08 schrieb Mike Horvath:
> Forgot to mention there are strange square shaped artifacts in the blue
> end caps of the scene. They exist in the source images as well. I don't
> know what causes them.
Looks like radiosity artifacts to me.
Your scene may need a more exhaustive radiosity pretrace.
Make sure you have `always_sample` set to `off` (or not set at all).
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 4/3/2017 8:13 AM, clipka wrote:
> Am 03.04.2017 um 03:08 schrieb Mike Horvath:
>
>> Forgot to mention there are strange square shaped artifacts in the blue
>> end caps of the scene. They exist in the source images as well. I don't
>> know what causes them.
>
> Looks like radiosity artifacts to me.
> Your scene may need a more exhaustive radiosity pretrace.
>
> Make sure you have `always_sample` set to `off` (or not set at all).
>
Here are my global settings. Radiosity is disabled.
global_settings
{
charset utf8
assumed_gamma 1.0
#if (!NoRadiosity)
ambient_light 0
radiosity
{
pretrace_start 0.08
pretrace_end 0.02
count 20
error_bound 1
recursion_limit 1
normal on
brightness 0.8
always_sample no
gray_threshold 0.8
media on
}
#end
#switch (TexQual)
#case (-1)
max_trace_level 4
#break
#case (0)
max_trace_level 4
#break
#case (1)
max_trace_level 4
#break
#case (2)
max_trace_level 4
#break
#end
}
Mike
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 4/3/2017 4:50 AM, omniverse wrote:
> Mike Horvath <mik### [at] gmailcom> wrote:
>> On 4/2/2017 8:52 PM, Mike Horvath wrote:
>>> I updated my spaceship scene and rendered new frames to be used with
>>> Pannellum panorama viewer.
>>>
>>> http://isometricland.net/panorama/pannellum.htm?config=pano-gh-scene-spinner.json
>>>
>>>
>>> I still was not able to turn the interior atmosphere due to render
>>> times. Even on the 16-core MS Azure server I am using for my renders
>>> this month, things were taking too long.
>>
>> Forgot to mention there are strange square shaped artifacts in the blue
>> end caps of the scene. They exist in the source images as well. I don't
>> know what causes them.
>
> Impressive scene. And maybe disorienting to some earthbound people. :)
>
Thanks!
> If able to see it, is the atmosphere both ground-hugging and cloudlets floating
> around?
>
No, it is a constant density.
> Wild guess about the squares. Could it be something about the ends being
> directly opposite and slightly reflective, max_trace_level?
> Although I don't recall seeing unusual squared patterns from anything like that
> before.
>
It could be, since I have it set very low.
Mike
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 03.04.2017 um 18:36 schrieb Mike Horvath:
> On 4/3/2017 8:13 AM, clipka wrote:
>> Am 03.04.2017 um 03:08 schrieb Mike Horvath:
>>
>>> Forgot to mention there are strange square shaped artifacts in the blue
>>> end caps of the scene. They exist in the source images as well. I don't
>>> know what causes them.
>>
>> Looks like radiosity artifacts to me.
>> Your scene may need a more exhaustive radiosity pretrace.
>>
>> Make sure you have `always_sample` set to `off` (or not set at all).
>>
>
> Here are my global settings. Radiosity is disabled.
Then I have no idea what might cause the artifacts.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Mike Horvath <mik### [at] gmailcom> wrote:
> On 4/3/2017 4:50 AM, omniverse wrote:
>
> > If able to see it, is the atmosphere both ground-hugging and cloudlets floating
> > around?
> >
>
> No, it is a constant density.
Ah, okay, I was using my imagination anyway. No I'm wondering what that would do
in the spiral light... if that is a light.
> > Wild guess about the squares. Could it be something about the ends being
> > directly opposite and slightly reflective, max_trace_level?
> > Although I don't recall seeing unusual squared patterns from anything like that
> > before.
> >
>
> It could be, since I have it set very low.
What is really strange about it, to me, is how one of the nearby artifacts
(after moving scene around) appears to have a 90 degree angle going on.
The square shapes form a sort of floor and wall, if thinking of it as partially
covering tiles without seams where they meet.
Hopefully you can see or find that too.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Mike Horvath <mik### [at] gmailcom> wrote:
>
> Here are my global settings. Radiosity is disabled.
>
>
> global_settings
> {
> charset utf8
.....
[snip]
Although I haven't seen your animation yet, the charset utf entry caught my
eye. Wasn't there another recent post about artefacts due to that? I think it
was in the post about the "Dokken" logo not rendering correctly.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 03.04.2017 um 21:47 schrieb omniverse:
> What is really strange about it, to me, is how one of the nearby artifacts
> (after moving scene around) appears to have a 90 degree angle going on.
> The square shapes form a sort of floor and wall, if thinking of it as partially
> covering tiles without seams where they meet.
> Hopefully you can see or find that too.
My guess is that the 360 degree image format is using a cube model, and
what you see is the edge of that box.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 04.04.2017 um 11:59 schrieb Kenneth:
> Mike Horvath <mik### [at] gmailcom> wrote:
>
>>
>> Here are my global settings. Radiosity is disabled.
>>
>>
>> global_settings
>> {
>> charset utf8
> ......
> [snip]
>
> Although I haven't seen your animation yet, the charset utf entry caught my
> eye. Wasn't there another recent post about artefacts due to that? I think it
> was in the post about the "Dokken" logo not rendering correctly.
"charset utf8" can't give you genuine render artifacts. It can only give
you wrong character shapes in a `text` object.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 4/4/2017 9:25 AM, clipka wrote:
> Am 03.04.2017 um 21:47 schrieb omniverse:
>
>> What is really strange about it, to me, is how one of the nearby artifacts
>> (after moving scene around) appears to have a 90 degree angle going on.
>> The square shapes form a sort of floor and wall, if thinking of it as partially
>> covering tiles without seams where they meet.
>> Hopefully you can see or find that too.
>
> My guess is that the 360 degree image format is using a cube model, and
> what you see is the edge of that box.
>
Yes, it uses 6 images to form a cube map. The artifacts are distributed
randomly across each of the 6 images.
Mike
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 4/2/2017 9:08 PM, Mike Horvath wrote:
> On 4/2/2017 8:52 PM, Mike Horvath wrote:
>> I updated my spaceship scene and rendered new frames to be used with
>> Pannellum panorama viewer.
>>
>> http://isometricland.net/panorama/pannellum.htm?config=pano-gh-scene-spinner.json
>>
>>
>>
>> I still was not able to turn the interior atmosphere due to render
>> times. Even on the 16-core MS Azure server I am using for my renders
>> this month, things were taking too long.
>>
>> Enjoy.
>>
>>
>> Mike
>
> Forgot to mention there are strange square shaped artifacts in the blue
> end caps of the scene. They exist in the source images as well. I don't
> know what causes them.
>
>
> Mike
Forgot to say I think the artifacts are the same size as the "render
blocks" (or whatever they're called) when rendering any scene. Sort of
like the mosaic preview.
Mike
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 4/2/2017 8:52 PM, Mike Horvath wrote:
> I updated my spaceship scene and rendered new frames to be used with
> Pannellum panorama viewer.
>
> http://isometricland.net/panorama/pannellum.htm?config=pano-gh-scene-spinner.json
>
>
> I still was not able to turn the interior atmosphere due to render
> times. Even on the 16-core MS Azure server I am using for my renders
> this month, things were taking too long.
>
> Enjoy.
>
>
> Mike
I rendered a higher resolution version of the scene.
http://isometricland.net/panorama/pannellum.htm?config=pano-gh-scene-spinner.json
It includes some but not all of the fixes I have been working on over
the last few days. I will probably do another render sometime soon,
though not at the same high resolution because it takes so long.
Mike
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 4/6/2017 11:16 PM, Mike Horvath wrote:
> On 4/2/2017 8:52 PM, Mike Horvath wrote:
>> I updated my spaceship scene and rendered new frames to be used with
>> Pannellum panorama viewer.
>>
>> http://isometricland.net/panorama/pannellum.htm?config=pano-gh-scene-spinner.json
>>
>>
>>
>> I still was not able to turn the interior atmosphere due to render
>> times. Even on the 16-core MS Azure server I am using for my renders
>> this month, things were taking too long.
>>
>> Enjoy.
>>
>>
>> Mike
>
> I rendered a higher resolution version of the scene.
>
> http://isometricland.net/panorama/pannellum.htm?config=pano-gh-scene-spinner.json
>
>
> It includes some but not all of the fixes I have been working on over
> the last few days. I will probably do another render sometime soon,
> though not at the same high resolution because it takes so long.
>
>
Okay I'll bite. How long did it take?
It is really good.
What's puzzling me is what looks like a shadow along the three arms of
the support.
--
Regards
Stephen
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 4/6/2017 6:35 PM, Stephen wrote:
> On 4/6/2017 11:16 PM, Mike Horvath wrote:
>> It includes some but not all of the fixes I have been working on over
>> the last few days. I will probably do another render sometime soon,
>> though not at the same high resolution because it takes so long.
>>
>>
>
> Okay I'll bite. How long did it take?
>
It took 2-3 days on a rented Microsoft Azure server. On my home PC it
would have taken 7-10 days I think. That's without the atmospheric media
I wanted to use in the scene, or the radiosity.
> It is really good.
>
Thanks! I have been working on it since ~2010 I think. Maybe earlier.
> What's puzzling me is what looks like a shadow along the three arms of
> the support.
>
I don't know either.
Mike
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 4/7/2017 1:06 AM, Mike Horvath wrote:
> On 4/6/2017 6:35 PM, Stephen wrote:
>> On 4/6/2017 11:16 PM, Mike Horvath wrote:
>>> It includes some but not all of the fixes I have been working on over
>>> the last few days. I will probably do another render sometime soon,
>>> though not at the same high resolution because it takes so long.
>>>
>>>
>>
>> Okay I'll bite. How long did it take?
>>
>
> It took 2-3 days on a rented Microsoft Azure server. On my home PC it
> would have taken 7-10 days I think. That's without the atmospheric media
> I wanted to use in the scene, or the radiosity.
>
Ah! that's not CPU ticks I hear. It is the sound of cents draining out
of your pocket. ;)
It was worth it. :)
>
>> It is really good.
>>
>
> Thanks! I have been working on it since ~2010 I think. Maybe earlier.
>
>
I remember the early posts and versions. Then you went quiet.
>> What's puzzling me is what looks like a shadow along the three arms of
>> the support.
>>
>
> I don't know either.
>
The strange thing is it looks familiar. I used a similar shape for a
flywheel, years ago.
--
Regards
Stephen
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 4/6/2017 8:34 PM, Stephen wrote:
> On 4/7/2017 1:06 AM, Mike Horvath wrote:
>> On 4/6/2017 6:35 PM, Stephen wrote:
>>> On 4/6/2017 11:16 PM, Mike Horvath wrote:
>>>> It includes some but not all of the fixes I have been working on over
>>>> the last few days. I will probably do another render sometime soon,
>>>> though not at the same high resolution because it takes so long.
>>>>
>>>>
>>>
>>> Okay I'll bite. How long did it take?
>>>
>>
>> It took 2-3 days on a rented Microsoft Azure server. On my home PC it
>> would have taken 7-10 days I think. That's without the atmospheric media
>> I wanted to use in the scene, or the radiosity.
>>
>
>
> Ah! that's not CPU ticks I hear. It is the sound of cents draining out
> of your pocket. ;)
>
> It was worth it. :)
>
Well, I am using the free trial period/balance you get when you create
an account. You get $200 which you can spend in 30 days. I have almost
used all $200.
Google offers $300 that you can spend in 12 months, as well as a
discount if you use Linux instead of Windows. I will be doing that next.
>>
>>> It is really good.
>>>
>>
>> Thanks! I have been working on it since ~2010 I think. Maybe earlier.
>>
>>
>
> I remember the early posts and versions. Then you went quiet.
>
>
I put it aside as render times were getting too long using just my home
PC. I am impatient.
>>> What's puzzling me is what looks like a shadow along the three arms of
>>> the support.
>>>
>>
>> I don't know either.
>>
>
> The strange thing is it looks familiar. I used a similar shape for a
> flywheel, years ago.
>
The "caps" on each end of the habitat are actually a little concave,
too, like a dish. Maybe that is influencing the render?
Mike
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 4/7/2017 3:51 AM, Mike Horvath wrote:
> On 4/6/2017 8:34 PM, Stephen wrote:
>> On 4/7/2017 1:06 AM, Mike Horvath wrote:
>>> On 4/6/2017 6:35 PM, Stephen wrote:
>>>> On 4/6/2017 11:16 PM, Mike Horvath wrote:
>>>>> It includes some but not all of the fixes I have been working on over
>>>>> the last few days. I will probably do another render sometime soon,
>>>>> though not at the same high resolution because it takes so long.
>>>>>
>>>>>
>>>>
>>>> Okay I'll bite. How long did it take?
>>>>
>>>
>>> It took 2-3 days on a rented Microsoft Azure server. On my home PC it
>>> would have taken 7-10 days I think. That's without the atmospheric media
>>> I wanted to use in the scene, or the radiosity.
>>>
>>
>>
>> Ah! that's not CPU ticks I hear. It is the sound of cents draining out
>> of your pocket. ;)
>>
>> It was worth it. :)
>>
>
> Well, I am using the free trial period/balance you get when you create
> an account. You get $200 which you can spend in 30 days. I have almost
> used all $200.
>
> Google offers $300 that you can spend in 12 months, as well as a
> discount if you use Linux instead of Windows. I will be doing that next.
>
It is certainly cheaper and tidier than buying hardware.
>>>
>>>> It is really good.
>>>>
>>>
>>> Thanks! I have been working on it since ~2010 I think. Maybe earlier.
>>>
>>>
>>
>> I remember the early posts and versions. Then you went quiet.
>>
>>
>
> I put it aside as render times were getting too long using just my home
> PC. I am impatient.
>
>
I've lost count of the projects I've abandoned for that reason.
>>>> What's puzzling me is what looks like a shadow along the three arms of
>>>> the support.
>>>>
>>>
>>> I don't know either.
>>>
>>
>> The strange thing is it looks familiar. I used a similar shape for a
>> flywheel, years ago.
>>
>
> The "caps" on each end of the habitat are actually a little concave,
> too, like a dish. Maybe that is influencing the render?
>
>
I've had a look at some of my old files and think I must be remembering
one of your original posts that puzzled me. (At least my long term
memory is still working. ;) )
--
Regards
Stephen
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Mike Horvath <mik### [at] gmailcom> wrote:
> I rendered a higher resolution version of the scene.
This is really nice - and really the only way to look at scenes like this!
Have you tried rendering it from 'ground level', as an inhabitant would see it?
Bill
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 4/7/2017 6:09 AM, Stephen wrote:
> On 4/7/2017 3:51 AM, Mike Horvath wrote:
>> On 4/6/2017 8:34 PM, Stephen wrote:
>>> On 4/7/2017 1:06 AM, Mike Horvath wrote:
>>>> On 4/6/2017 6:35 PM, Stephen wrote:
>>>>> On 4/6/2017 11:16 PM, Mike Horvath wrote:
>>>>>> It includes some but not all of the fixes I have been working on over
>>>>>> the last few days. I will probably do another render sometime soon,
>>>>>> though not at the same high resolution because it takes so long.
>>>>>>
>>>>>>
>>>>>
>>>>> Okay I'll bite. How long did it take?
>>>>>
>>>>
>>>> It took 2-3 days on a rented Microsoft Azure server. On my home PC it
>>>> would have taken 7-10 days I think. That's without the atmospheric
>>>> media
>>>> I wanted to use in the scene, or the radiosity.
>>>>
>>>
>>>
>>> Ah! that's not CPU ticks I hear. It is the sound of cents draining out
>>> of your pocket. ;)
>>>
>>> It was worth it. :)
>>>
>>
>> Well, I am using the free trial period/balance you get when you create
>> an account. You get $200 which you can spend in 30 days. I have almost
>> used all $200.
>>
>> Google offers $300 that you can spend in 12 months, as well as a
>> discount if you use Linux instead of Windows. I will be doing that next.
>>
>
>
> It is certainly cheaper and tidier than buying hardware.
>
>
>>>>
>>>>> It is really good.
>>>>>
>>>>
>>>> Thanks! I have been working on it since ~2010 I think. Maybe earlier.
>>>>
>>>>
>>>
>>> I remember the early posts and versions. Then you went quiet.
>>>
>>>
>>
>> I put it aside as render times were getting too long using just my home
>> PC. I am impatient.
>>
>>
>
> I've lost count of the projects I've abandoned for that reason.
>
>
>>>>> What's puzzling me is what looks like a shadow along the three arms of
>>>>> the support.
>>>>>
>>>>
>>>> I don't know either.
>>>>
>>>
>>> The strange thing is it looks familiar. I used a similar shape for a
>>> flywheel, years ago.
>>>
>>
>> The "caps" on each end of the habitat are actually a little concave,
>> too, like a dish. Maybe that is influencing the render?
>>
>>
>
> I've had a look at some of my old files and think I must be remembering
> one of your original posts that puzzled me. (At least my long term
> memory is still working. ;) )
>
>
Source code is available here:
http://www.mediafire.com/file/gp2q0br1737h9du/GH_LoadingScreen_src_v2_9_0.zip
It is in the "gh_scene_spinner" directory. The file name is
"gh_scene_spinner_interior_b.pov". Good luck tracking down all the
dependencies however!
Mike
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 4/7/2017 8:25 AM, Bill Pragnell wrote:
> Mike Horvath <mik### [at] gmailcom> wrote:
>> I rendered a higher resolution version of the scene.
>
> This is really nice - and really the only way to look at scenes like this!
>
> Have you tried rendering it from 'ground level', as an inhabitant would see it?
>
> Bill
>
>
I'm afraid there won't be much to see at ground level except some very
primitive looking people models.
Mike
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
